Buchardt S400 Speaker Review

[MZKM]

Have you given them some break-in time before measuring? Buchardt recommends 100 hours minimum which I found a bit extreme but in fact, these are the only pair of speakers in my memory (apart from a few subs) that actually did require break-in to sound even remotely normal, not to say great. At least 10-12 hours are necessary since out of the box their woofer is simply not working at all, as if it is stuck. This is by design, so they say in the documentation.

Here is the quote:

How long does it take to break them in?

100 hours will get you most of the way there.

Why is this important? Well the S400 uses very large rubber surrounds on both the main woofer and the passive radiator. The rubber is pretty stiff out of the box and needs “massage” to soften up and thereby be easier to drive to simplify it. This will give a better bass performance and makes the integration to the tweeter better as well. You can expect an pretty dramatic change in sound throughout the hours you put in to them.

 

[617]

Instructive to see high resolution midbass band measurements of that woofer; sb acoustics has driven breakup up quite high but I suspect the gremlin at 500hz is woofer related.

The cabinets bracing is somewhat compromised by the passive radiator, which prevents a brace from tying to the back. Perhaps it doesn't matter but I would expect a more inert cabinet at this price. The PR is a costly design element so I would expect bass performance better than with a port.

This is a good design and the aesthetics are very strong, but I would have liked to have seen a more harmonious directivity marriage between the drivers.

Big points for the design and 10 year warranty though, as well as the fact the designer comments here. I'd like to hear them if I'm ever in the market for speakers again.

 

[astr0b0y]

Only 3 hrs needed for DNA repair via B400! Seems a paltry price to pay compared to 100 hours break in.

But seriously, I had looked at Buchardt (how do you pronounce that?) as a possible speaker to try, probably their active model. But seeing that they market break in time has ruled them out completely.

 

[napilopez]

Thanks @amirm!

I measured this speaker a while back, and I enjoyed it quite a bit. Some flaws, but it seems to do well on your klippel too other than that (imo minor) resonance.

Back then, I was only measuring in 15 degree intervals and doing incomplete rear measurements, but I do think it can be interesting to compare the data. In some respects, my measurements are closer to the company results, although mine have a slight downward tilt.

As a casual observation, I do note once again that a speaker with narrower horizontal directivity did not excite you as much as the wider directivity ones. Though the S400 seems a bit wider than the usual big waveguide speaker, perhaps because of that little tweeter.

Anyway, for comparison, here's my spin based on 15-degree intervals, measured between tweeter and waveguide:

While the crossover dip is present in the bigger averaged curves, I did not get the same crossover big dip in the horizontal data. My polar looks closer to the company's; there's a slight scoop but not bad:

Also visible is the SPL plot, where the bit of off-axis energy is flattened out off axis.

I noted in my vertical measurements that the S400 is extremely sensitive to vertical positioning within the listening window. I think the most sensitive of any speaker I've tested. Here's the speaker at 0/5/10/15/30 degrees vertical.

Look at the difference happening just 5 degrees below the stated reference axis, for instance:

It instead clears up significantly at steeper vertical angles:

I liked it, but noticed I did not like it in the nearflied, I suspect because of that finicky vertical positioning. In the Fairfield I did not have that problem, although I took care to align the reference axis properly.

I'd still consider these great results in the grand scheme of things for a small internet-direct personally, but your mileage may vary.

 

[Speaker-Inquisitor]

If breaking in is really important, i would suspect the manufacturer sends an already broken in sample for reviewing purposes.

 

[Bamyasi]

If breaking in is really important, i would suspect the manufacturer sends an already broken in sample for reviewing purposes.

That's the question and I would love to hear responses from Amir and from Mad as well. Once again, personally I am not a fan of this "break-in" business either but in this particular case there seems to be something in the design that actually does require at least some amount.

 

[MZKM]

If breaking in is really important, i would suspect the manufacturer sends an already broken in sample for reviewing purposes.

”You don’t buy cars with their engines already broken-in.”
A common reply.

 

[bigjacko]

@napilopez How did you measure speakers? Can you share what things did you use? Thank you.

 

[jonfitch]

The woofer looks like the woofer used on the Performa3 series from Revel? Its only 6" though?

I think this is a Scanspeak woofer, whereas Revel uses SB Acoustics.

 

[thewas]

Buchardt to my knowledge also use SB Acoustics drivers as some of their people started as danish SB distributors.

 

[napilopez]

@napilopez How did you measure speakers? Can you share what things did you use? Thank you.

How much detail do you want? Hahah. I use the usual quasi-anechoic method and nearfield splicing, largely explained in this Jeff Bagby whitepaper(though I use REW instead). This course helped a lot too. There are flaws with this method, but it's good for a general overview. The biggest flaw is the lack of resolution for the lower half of frequencies, but it can still be pretty informative with some effort and combined with measurements at various distances.

Gear/setup: it's REW with a MiniDSP Umik-1 that's been calibrated by CSL for extra accuract. I put the speaker stand on a turntable/lazy Susan I labelled with various angles. It's important to make sure the rotation axis is aligned with the front baffle of the speaker.

I place the setup on my kitchen island, elevating the speaker about 6 feet. I then measure and repeat for each angle. For vertical, I flip the speaker on its side and repeat, making sure to keep the same reference axis. Flipping the speaker usually yields slight differences, I think reflections from the stand, but not too bad for averaged data.

I then gate the responses at 6.5ms, a little before the first reflection hits in my room. If I import all the measurements into vituixcad, it then automatically generates a spinorama, which i can export back into REW. I then splice the nearfield data as detailed in the Bagby paper for the bass and voilà.

 

[MZKM]

Buchardt to my knowledge also use SB Acoustics drivers as some of their people started as danish SB distributors.

Yeah, the design looks like SB (has the line dashes).

 

[jonfitch]

It's interesting there's a dip at the crossover frequency with these speakers on the horizontal spins. I would have thought the purpose of building such a giant waveguide is to be able to lower the crossover frequency sufficiently to avoid a crossover dip entirely.

 

[MZKM]

Have you given them some break-in time before measuring?

Buchardt's own measurements match, and you would assume if they state burn-in improves the performance (especially in the bass), that they would have used a broken-in speaker for measurements.

 

[MZKM]

It's interesting there's a dip at the crossover frequency with these speakers on the horizontal spins. I would have thought the purpose of building such a giant waveguide is to be able to lower the crossover frequency sufficiently to avoid a crossover dip entirely.

They used a <1" soft dome, so that likely is the lowest they could have gone (why they didn't use a 1", I don't know).

 

[thewas]

(why they didn't use a 1", I don't know).

According to them they used a 3/4" one to keep the directivity constantly wide at very high frequencies.

 

[jonfitch]

They used a <1" soft dome, so that likely is the lowest they could have gone (why they didn't use a 1", I don't know).

Yeah, they even said they tried using the Satori TW29BN and found no benefit, which makes no sense to me considering that same tweeter is used in some other speakers with a ~1700hz crossover, which definitely would have eliminated that dip here.

 

[Maplegrovemusic]

”You don’t buy cars with their engines already broken-in.”
A common reply.

I believe engines are tested and run to make sure they perform before being installed in a new car . They use a special viscosity oil different from conventional oil for that . Then in most new car manuals they give recomendations on how to break in your engine for a set amount of miles .

 

[thewas]

In the end it depends of course also on the waveguide geometry and breakup of the dome, but generally a smaller dome or cone leads to a wider directivity which is a pure geometrical thing.

 

[napilopez]

The prominence of the resonance at 600Hz measurements in Amir's data is interesting - I wonder if that's one area of unit variation. My quasi anechoic measurements probably don't have enough resolution this area to give a reliable picture of what's going on, but I did try to see if I could find evidence of it in the old data of my sample.

There seems to be broad lump around 600Hz in my measurements off-axis. For those of you with speaker design experience - would that 600Hz resonance show up in nearfield measurements of the radiator/woofer? Or would it not be apparent if it's from the cabinet?

In any case, here's what I get for the individual woofer and radiator responses, as well as the baffle step-corrected sum of their responses (graphs not level matched):

You can see the same scoop that shows up from 500-800 Hz as Buchardt, although not the lil peak:


Something going on around that area regardless, but it definitely look worse in Amir's case.

My CSDs for the curious:

Woofer

Quasi-anechoic